Re: pixel fonts

On Aug 8,  7:36pm, lee@sq.com wrote:

> Bill Hill <billhill@microsoft.com> wrote:
> > I'd go further. A BIG step backward!

> I think I agree, but I also think that if the HTTP proposal for including
> client resolution, geometry and colour informatin is included, bitmap
> fonts could work.


Provided the server was able to generate the requested fonts at the requested
color and resolution. This would impact latency a little, and would also impact
cacheing performance somewhat.

Treating a pixel font (or an outline font rendered with anti-aliasing) as
an alpha channel rather than as image data has interesting possibilities,
such as true antialiased text of any color on any color background, or
foreground images on display type.

> The main advantages I see over using individual gifs
> are sending multiple images in a single transaction (like HTTP 1.1)

Yes, the keepalive could be used for that. Then again, one could define a
pixelfont format that puts all the images in one file, either using a format
that can contain multiple independent pictures, using a large image with the
letters aranged on a grid, or even using a MIME multipart to batch up the
individual letters.

> and
> in positioning to align baselines with surrounding text.  The latter
> consideration seems to be the more important of the two.

It is interesting to study the proposed alIG (alignment) chunk for PNG,
which seems to cater for precisely this:

http://www.hensa.ac.uk/png/png-group/documents/png-proposed-chunks-0.960710.html


> I think that the requests to use bitmap fonts are coming from font
> designers who don't want to allow their outline fonts to be embedded, for
> fear of copyright infringements and piracy.  This is a valid concern
> which OpenType must address.

Not just Opentype, any font which is used on the Web must adress these
concerns. There is a balance to be struck between hindering legitimate use and
hindering illegitimate use; just a s there is a balance to be struck by font
vendors between focussing on a small number of phototypsetting bureaux and a
large number of Web site authors as potential markets.


> There are also concerns that a hand-tuned bitmap font can give a better
> appearance than an outline font at low resolutions.  I think that this is
> utter nonsense unless it can be combined with knowledge of the client's
> screen size, resolution and number of colours.  Armed with that knowledge,
> though, it is possibly true.

Both bitmap and outline fonts can be rendered badly, or rendered well. I agree
that bitmaps fonts require extra information in the request (display density,
primarily, plus whether it really is monochrome or whether it has contone
capability (greyscale or color)).

>   However, it would be necessary to show how
> the bitmap font approach can be made amenable to high quality printing.

That seems to be an important requirement, as print-on-demand shops that
generate 'books" of printed and bound document sets looks like a growth area.

> So a formal write-up is needed, including a technical specification as
> to exactly how it would be implemented.  But is it worth anyone spending
> a few days doing that?  Would anyone listen?

If a proposal for pixel-based fonts was presented which addressed these various
issues and was technically sound, I for one would certainly listen. I am the
chair of the W3C Fonts Working Group, by the way.



-- 
Chris Lilley, W3C                          [ http://www.w3.org/ ]
Graphics and Fonts Guy            The World Wide Web Consortium
http://www.w3.org/people/chris/              INRIA,  Projet W3C
chris@w3.org                       2004 Rt des Lucioles / BP 93
+33 93 65 79 87            06902 Sophia Antipolis Cedex, France

Received on Friday, 9 August 1996 07:44:46 UTC